PSX circuit breaker not tripping, booster trips instead

NScaleKen Nov 6, 2018

  1. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    I have a new PSX4 unit. During the first initial install and test 1-3 passed and caught a short, then the 4th did not catch the short and the booster tripped. After that first test all 4 units on the PSX4 failed repeatedly at catching a short doing the units initial testing routine (which is simply connect it normally and short it at the output terminal). Not sure what to make of 3 working as described initially then not working, with absolutely nothing changed or even touched.

    To try and fix it I messed with the devices CV values to adjust the trip current, at setting 1 of around 1.27 amps one unit caught a short once, then continued to let the short pass through to the booster which tripped. I then adjusted the DCC++ currentmonitor.h settings from 300 to 750 which should allow around 2 amps which I am not sure applies to short circuit sensing or just track power? Still the arduino/pololu control station is tripping not the PSX circuit breakers. I have tried resetting to factory several times between attempts at current trip value adjustments in the 1.27 to 5 amp range.

    Any advice?
     
  2. TrainboySD40

    TrainboySD40 TrainBoard Member

    257
    21
    22
    NScaleKen likes this.
  3. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    I did that, it did not help at first. However the latest update on it... I ate some dinner, turned system back on with one PSX unit of the 4 connected, and tried CV 49 to value 1 for either the 4th or 5th time, just to try, and it worked on the one unit I had hooked up.

    I swear I tried it at least 4 times in that same configuration, boogled, I hooked up the other 3 and without changing anything simply clamping the wires from 3 daisy chained units into the one connected to the main power and I tried them with a CV 49 value at 2 which was the last attempt before dinner, and they worked. the 4th unit works now which never caught a short before at all.

    So I guess dinner works? I had 4 cheese quesadillas in corn tortillas, and 3 chicken and green chile tamales. I think it might have been the quesadillas made with corn tortillas that was the crucial step.

    or something, jesus I have no idea at this point and while it is working now, for $130 it seems pretty darn unreliable. Going to call the shop I used tomorrow and ask about it. See what they cook to make the darn things work :p
     
    drbnc likes this.
  4. RBrodzinsky

    RBrodzinsky November 18, 2022 Staff Member TrainBoard Supporter In Memoriam

    5,685
    2,780
    98
    Ken, I have always found the CV49=1 setting (1.2 amps) a bit flakey. The CV49=2 2.5 amp trip point is much more reliable. But, you need to make sure your booster is putting out at least 3 amps for that trip current to work. The booster current needs to be at least 25% greater than the trip point for reliability.

    But really, it was the green chile tamales.
     
    NScaleKen likes this.
  5. DCESharkman

    DCESharkman TrainBoard Member

    4,396
    3,025
    87
    They are very reliable if setup correctly and that the booster power is high enough.

    I have dozens of them and never had any issues at all.
     
    NScaleKen likes this.
  6. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    It may be booster power thanks for the advice on that, I will research DCC++ track power more. If I have more locomotives on the track operating would that help test this? I have been running none or just one in testing, and its catching shorts maybe 1 in 10 times. I think there is an isolation issue with the layout, the Kato double crossover is possibly letting current through, when I thought it isolated the 4 tracks coming into it as one needs them all powered to get across the center. I tested a subdistrict leading into it and it tripped the PSX on the subdistrict on the other side of the Kato double crossover, both PSX units worked and caught the short that time, it was applied at the PSX output terminal for one track leading into the double crossover and tripped the PSX sending power to the subdistrict on the other side of the double crossover. I guess I need that switch totally contained in one subdistrict?
     
  7. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    question, I have a Meanwell NES-350-12 0-29amp power supply feeding the pololu mc33926 motor driver shield, the mc33926 is rated to deliver just under 3 amps continuously to a channel. I have no idea what it is feeding the track though as I do not have a ramp meter device. How does that work with this DCC++ system? the locomotives and anything drawing power from the main track power create the draw, the power supply tries to deliver it, and the motor shield is a choke point that modifies the power waveform and limits the amount the power supply can deliver? the mc33926 can go to 5 amps in short peaks, is that its circuit protection limit and it will cut off like its a short on that channel after that? just trying to wrap my head around how this works as compared to other things that use motors and controllers like 3d printers versus the lightswitch operation of a DC system.
     
  8. RBrodzinsky

    RBrodzinsky November 18, 2022 Staff Member TrainBoard Supporter In Memoriam

    5,685
    2,780
    98
    Couple of things

    Kato double crossover: while the inner tracks are isolated by the crossover, the two long outside tracks are through tracks. there needs to be complete isolation of the section from other PSX sections.

    No, number of locos on the track will not change behavior. Just means you are starting out closer to the trip point. When you short a circuit, the current will spike upwards until the flow overpowers the most susceptible device on the circuit. At home, this is typically a 15amp circuit breaker. In this case, you are looking to see which breaker pops first, the PSX or the motor shield. Your maximum (without PSX) is the rating of the shield — doesn’t matter what the power supply into it can provide. The shield doesn’t put out a constant current, it allows the devices in the circuit to pull what they need, until the maximum current the shield can produce is exceeded.

    If the shield will trip at less than 3 amps, you need to set the trip current of the PSX to 1.2 amps (CV49=1). I would do a reset of the PSX and then set, as something else in the settings may have changed. The 2.5 amp setting is too close to your supply’s trip point.

    One suggestion, since you have Unitrack: set up two small sections which are not physically connected, one on the PSX, the other not. Make sure that shorting the PSX connected track trips it, and that signal is still going from command station to the other track.
     
    NScaleKen likes this.
  9. mikegillow

    mikegillow TrainBoard Member

    116
    117
    13
    NScaleKen likes this.
  10. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    Ah thanks for pointing out the outer track outside rail passes through Rick, I totally missed that in my haste to run trains :p

    The shield is good for 5 amps in peaks and 3 amps continuous power. The 2 setting should have enough overhead.

    I am pretty sure it's partly the power supplied as the DCC++ code was sensing at 800Mah, leading to the failures when testing before hooking up anything to the outputs, and partly the wiring once the PSX is on the layout which has messy at the central hub temporary mainline bus. Linked 2 PSX for sure through the double X, lord knows what else. I will redo the wiring rather than hunt down errors in a bowl of spaghetti.

    The boards still sometimes dont catch shorts, I can't really assume anything is wrong with them since my motor shield is possibly cutting the circuit first or not putting out enough amps.

    I have an ESU CabControl on backorder, 5amp power specificly for DCC, should solve some of the issues. I am trying to learn it all at once and the control and power is taking up way to much time from modeling landscapes and industries. Hope to get more creative time once I get the ESU system. DCC++ is great, will totally use it on a small logging railroad 2nd layout, but I want to learn how to make landscapes and industries first.

    Thanks for the link Mike, I had seen that a while ago but needed reminding.
     
  11. DCESharkman

    DCESharkman TrainBoard Member

    4,396
    3,025
    87
    It is always best when cutting your teeth on trains and DCC, to start with a good DCC controller. This way you learn a lot about the wiring etc for good DCC performance. You can still interface DCC++ elements (LDCC) as needed, and as you said start a second layout and now deal with the idiosyncrasies of DCC++.
     
    NScaleKen likes this.
  12. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    I ran into the back of a switch in the yard. I checked to see if I had switch power still, and I did. I switched the turnout and proceeded ahead. It works, at least in the areas I got wired correctly. Haven't fixed the double crossover yet as I never have shorts from it, since it can't create a dead end only a pass through.

    Many of the times I have tested the PSX previously it has not worked, but it may have been the arduino was not reset properly and I was using the 800Mah setting not the 2a setting, which it is now. Seems to be catching the stuff it was intended for, if not testing shorts directly at the output terminals which it still tends to miss and the motor shield catches.

    And to me more confusing, if I use 16 gauge wire to test short at the terminals it always fails, if I use a 24 gauge testing double ended alligator clip it passes quite often. Is the lesser resistance in the path short 'faster'? and the fast reaction pololu circuit protection gets it before the PSX, maybe?
     
  13. NScaleKen

    NScaleKen Permanently dispatched

    157
    128
    5
    Forgot to mention, I have a digitrax AR-1 analog relay reverser, behind one of the PSX units, and it is working so far every time. Hasn't been slower than the PSX so far. I had heard about that issue and was thinking of running it in front of the PSX units but thought I would try it out first. It seems to work, there may be a delay setting in the default firmware values now. Not sure, but it might explain the Pololu beating the PSX to the punch on shorts maybe, unless it doesn't work like that and I'm not thinking of that wiring setup correctly. In any case its catching shorts at switches and the AR-1 is beating it to reverse that loop even with the AR-1 powered from the output of a PSX.
     
  14. DCESharkman

    DCESharkman TrainBoard Member

    4,396
    3,025
    87
    Perhaps your should isolate that part of the layout and only use the AR1. And jumper around it connected to the PSX, or make life simple and get the PSX-AR to keep everything matched at the same trip settings. Consistency breeds success, and that works for using the same wire, the same connectors and the same power control circuitry.
     

Share This Page